Recherche avancée

Sur d’autres sites (5694)

• Anomalie #2542 : forum interne

  8 mars 2012, par christophe ulmer

  avec sqlite3 , cela fonctionne

• How to send encoded video (or audio) data from server to client in a way that's decodable by webcodecs API using minimal latency and data overhead

  11 janvier 2023, par Tiger Yang

  My question (read entire post for context) :

  


  Given the unique circumstance of only ever decoding data from a specifically-configured encoder, what is the best way I can send the encoded bitstream along with the bare minimum extra bytes required to properly configure the decoder on the client's end (including only things that change per stream, and omitting things that don't, such as resolution) ? I'm a sucker for zero compromises, and I think I am willing to design my own minimal container format to accomplish this.

  


  Context and problem :

  


  I'm working on a remote desktop implementation that consists of a server that captures and encodes the display and speakers using FFmpeg and forwards it via pipe to a go (language) program which sends it on two unidirectional webtransport streams to my client, which I plan to decode using the webcodecs API. According to MDN, the video decoder needs to be fed via .configure() an object containing the following : https://developer.mozilla.org/en-US/docs/Web/API/VideoDecoder/configure before it's able to decode anything.

  


  same goes for the audio decoder : https://developer.mozilla.org/en-US/docs/Web/API/AudioDecoder/configure

  


  What I've tried so far :

  


  Because this remote desktop will be for my personal use only, it would only ever receive streams from a specific encoder configured in a specific way encoding video at a specific resolution, framerate, color space, etc.. Therefore, I took my video capture FFmpeg command...

  


  videoString := []string{
        "ffmpeg",
        "-init_hw_device", "d3d11va",
        "-filter_complex", "ddagrab=video_size=1920x1080:framerate=60",
        "-vcodec", "hevc_nvenc",
        "-tune", "ll",
        "-preset", "p7",
        "-spatial_aq", "1",
        "-temporal_aq", "1",
        "-forced-idr", "1",
        "-rc", "cbr",
        "-b:v", "500K",
        "-no-scenecut", "1",
        "-g", "216000",
        "-f", "hevc", "-",
    }


  


  ...and instructed it to write to an mp4 file instead of outputting to pipe, and then I had this webcodecs demo https://w3c.github.io/webcodecs/samples/video-decode-display/ demux it using mp4box.js. Knowing that the demo outputs a proper .configure() object, I blindly copied it and had my client configure using that every time. Sadly, it didn't work, and I since noticed that the "description" part of the configure object changes despite the encoder and parameters being the same.

  


  I knew that mp4 files worked via mp4box, but they can't be streamed with low latency over a network, and additionally, ffmpeg's -f parameters specifies the muxer to use, but there are so many different types.

  


  At this point, I think I'm completely out of my depth, so :

  


  Given the unique circumstance of only ever decoding data from a specifically-configured encoder, what is the best way I can send the encoded bitstream along with the bare minimum extra bytes required to properly configure the decoder on the client's end (including only things that change per stream, and omitting things that don't, such as resolution) ? I'm a sucker for zero compromises, and I think I am willing to design my own minimal container format to accomplish this. (copied above)

  


• Minimal sample of muxing two streams with no reencoding (av_interleaved_write_frame fails)

  19 juillet 2022, par Alvein

  What I'm trying to do : having two files, one is video-only and the other is audio-only, with identical durations, I want to "join" them in a single container.

  


  I previously made a routine which just copied all the streams inside a container to another one. No reencoding, etc. This works perfectly :

  


  while(true) {
    pkIn=av_packet_alloc();
    if(NULL==pkIn) {
        fprintf(stderr,"av_packet_alloc() failed");
        break;
    }
    iError=av_read_frame(fcIn,pkIn);
    if(0>iError)
        if(AVERROR_EOF==iError)
            break;
        else {
            fprintf(stderr,"av_read_frame() failed");
            break;
        }
    stIn=fcIn->streams[pkIn->stream_index];
    stOut=fcOut->streams[pkIn->stream_index];
    log_packet(fcIn,pkIn,"in");
    av_packet_rescale_ts(pkIn,stIn->time_base,stOut->time_base);
    pkIn->pos=-1;
    log_packet(fcOut,pkIn,"out");
    iError=av_interleaved_write_frame(fcOut,pkIn);
    if(0>iError) {
        fprintf(stderr,"av_interleaved_write_frame() failed");
        break;
    }
    av_packet_free(&pkIn);
}


  


  I just did the analogy and tried to do the same, but taking each stream from a distinct container, like this :

  


  while(true) {
    if(!bVideoInEOF) {
        pkVideoIn=av_packet_alloc();
        if(NULL==pkVideoIn) {
            fprintf(stderr,"av_packet_alloc(video in) failed");
            break;
        }
        iError=av_read_frame(fcVideoIn,pkVideoIn);
        if(0>iError)
            if(AVERROR_EOF==iError)
                bVideoInEOF=true;
            else {
                fprintf(stderr,"av_read_frame(video in) failed");
                break;
            }
        if(!bVideoInEOF) {
            log_packet(fcVideoIn,pkVideoIn,"video in");
            av_packet_rescale_ts(pkVideoIn,stVideoIn->time_base,stVideoOut->time_base);
            pkVideoIn->pos=-1;
            pkVideoIn->stream_index=stVideoOut->index; // Edit (2022-07-19)
            log_packet(fcVideoIn,pkVideoIn,"video out");
            iError=av_interleaved_write_frame(fcOut,pkVideoIn);
            if(0>iError) {
                fprintf(stderr,"av_interleaved_write_frame(video out) failed");
                break;
            }
        }
        av_packet_free(&pkVideoIn);
    }
    if(!bAudioInEOF) {
        pkAudioIn=av_packet_alloc();
        if(NULL==pkAudioIn) {
            fprintf(stderr,"av_packet_alloc(audio in) failed");
            break;
        }
        iError=av_read_frame(fcAudioIn,pkAudioIn);
        if(0>iError)
            if(AVERROR_EOF==iError)
                bAudioInEOF=true;
            else {
                fprintf(stderr,"av_read_frame(audio in) failed");
                break;
            }
        if(!bAudioInEOF) {
            log_packet(fcAudioIn,pkAudioIn,"audio in");
            av_packet_rescale_ts(pkAudioIn,stAudioIn->time_base,stAudioOut->time_base);
            pkAudioIn->pos=-1;
            pkAudioIn->stream_index=stAudioOut->index; // Edit (2022-07-19)
            log_packet(fcAudioIn,pkAudioIn,"audio out");
            iError=av_interleaved_write_frame(fcOut,pkAudioIn);
            if(0>iError) {
                fprintf(stderr,"av_interleaved_write_frame(audio out) failed");
                break;
            }
        }
        av_packet_free(&pkAudioIn);
    }
    if(bVideoInEOF&&bAudioInEOF)
        break;
}


  


  I know the previous code looks like redundant but I wanted to leave both streams "processing" separated the way you understand my plans.

  


  Anyway, that code ends quickly with "av_interleaved_write_frame(audio out) failed".

  


  The error detail is "Invalid argument", and the debugger shows this :

  


  


  Application provided invalid, non monotonically increasing dts to
muxer in stream 0.

  


  


  If I disable any of the main blocks "if(!bVideoInEOF)" / "if(!bAudioInEOF)", the file is written successfully, with the obvious lack of the disabled stream.

  


  I'm new into using this library so probably I'm doing something really stupid, or missing something obvious.

  


  Suggestions ?

  


  Edit (2022-07-19) :

  


  By checking the logs, I noticed I was writing every frame to the stream #0. Hence, the horrible jumps in PTS/DTS.

  


  Code edited by adding the corresponding "...->stream_index=" before each call to av_interleaved_write_frame().

  


  ...

  


  Though it works, I still think my code is far from perfect. Comments are welcome.